The formation of blood clots does not only limit bleeding in case of an injury (hemostasis), but may lead to serious organ damage and death in the context of atherosclerotic diseases by occlusion of an important artery or vein. Thrombosis is thus blood clot formation at the wrong time and place. It involves a cascade of complicated and regulated biochemical reactions between circulating blood proteins (coagulation factors), blood cells (in particular platelets) and elements of an injured vessel wall. Anticoagulation and antithrombotic treatment aim at inhibiting the formation of blood clots in order to prevent these dangerous consequences, such as myocardial infarction, stroke, loss of a limb by peripheral artery disease or pulmonary embolism. Given the importance of these diseases, it is rather surprising that antithrombotic therapy relied on a few drugs for many years, namely Aspirin to inhibit platelets, Heparin that indirectly inhibits the coagulation Factors IX, X and II (thrombin), and oral Warfarin that inhibits Vit K-dependent factors (VII, IX, X, II and Prot C). After some time, low molecular weight Heparins (inhibiting Factors X and II to various degrees) became anticoagulants of choice, largely because of their ease of application (once a day subcutaneous injection with no monitoring need). With growing understanding of the processes involved in thrombosis a growing number of specific inhibitors of coagulation factors have been developed. However, a better efficacy/safety ratio could to date not be obtained with them. Direct thrombin inhibitors, in particular, were linked to increased bleeding complications in large clinical trials.
Aspirin also provides a protective effect against thrombosis. It induces a long-lasting functional defect in platelets, detectable clinically as a prolongation of the bleeding time, through inhibition of the cyclooxygenase activity of the human platelet enzyme prostaglandin H-synthase (PGHS-1) with doses as low as 30 to 75 mg. Since gastrointestinal side effects of aspirin appear to be dose-dependent, and for secondary prevention, treatment with aspirin is recommended for an indefinite period, there are practical reasons to choose the lowest effective dose. Further it has been speculated that a low dose (30 mg daily) might be more anti-thrombotic but attempts to identify the optimal dosage have yielded conflicting results. It has been claimed that the dose of aspirin needed to suppress fully platelet aggregation may be higher in patients with cerebrovascular disease than in healthy subjects and may vary from time to time in the same patient. However, aspirin in any daily dose of 30 mg or higher reduces the risk of major vascular events by 20% at most, which leaves much room for improvement.
Further, the inhibiting role of aspirin may lead to prevention of thrombosis as well as to excess bleeding. The balance between the two depends critically on the absolute thrombotic versus hemorrhage risk of the patient.
In patients with acute myocardial infarction, reduction of infarct size, preservation of ventricular function and reduction in mortality has been demonstrated with various thrombolytic agents. However these agents still have significant shortcomings, including the need for large therapeutic doses, limited fibrin specificity, and significant associated bleeding tendency. Recombinant tissue plasminogen activator (t-PA) restores complete patency in just over one half of patients, whereas streptokinase achieves this goal in less than one third. Further, reocclusion after thrombolytic therapy occurs in 5 to 10% of cases during the hospital stay and in up to 30% within the first year according to Verheugt et al., J. Am. Coll. Cardiol. (1996) 27:618-627. Thus numerous studies have examined the effects of adjunctive antithrombin therapy for patients with acute myocardial infarction. As an example, U.S. Pat. No. 5,589,173 discloses a method for dissolving and preventing reformation of an occluding thrombus comprising administering a tissue factor protein antagonist, which may be a monoclonal or polyclonal antibody, in adjunction to a thrombolytic agent. Monoclonal antibodies have already been shown to be of therapeutic value as antithrombotic agents. The first approved drug in this field was Abciximab (ReoPro™), a humanized Fab fragment of a murine monoclonal antibody (7E3) against platelet GP IIbIIIa receptors. Murine antibodies have characteristics which may severely limit their use in human therapy. As foreign proteins, they may elicit an anti-immunoglobulin response termed human anti-mouse antibody (HAMA) that reduces or destroys their therapeutic efficacy and/or provokes allergic or hypersensitivity reactions in patients, as taught by Jaffers et al., Transplantation (1986) 41:572. The need for readministration in therapies of thromboembolic disorders increases the likelihood of such immune reactions. While the use of human monoclonal antibodies would address this limitation, it has proven difficult to generate large amounts of such antibodies by conventional hybridoma technology.
Recombinant technology has therefore been used to construct “humanized” antibodies that maintain the high binding affinity of murine monoclonal antibodies but exhibit reduced immunogenicity in humans. In particular, chimeric antibodies have been suggested in which the variable region (V) of a non-human antibody is combined with the constant (C) region of a human antibody. As an example, the murine Fc fragment was removed from 7 E3 and replaced by the human constant immunoglobulin G Fab region to form a chimera known as c7 E3 Fab or abciximab. Methods of obtaining such chimerical immunoglobulins are described in detail in U.S. Pat. No. 5,770,198.
The potential for synergism between GPIIb/IIIa inhibition by monoclonal antibody 7 E3 Fab and thrombolytic therapy was evaluated by Kleiman et al., J. Am. Coll. Cardiol (1993) 22:381-389. Major bleeding was frequent in this study. Hence, the potential for life-threatening bleeding is clearly a major concern with this combination of powerful antithrombotic compounds.
Tissue Factor (TF), being a membrane glycoprotein functioning as a receptor for Factor VII and VIIa and thereby initiating the said extrinsic pathway, has been investigated as a target for anticoagulant therapy. In addition to this role, TF has been implicated in pathogenic conditions such as vascular disease and gram-negative septic shock. A study attempting to characterize the anticoagulant potential of murine monoclonal antibodies showed that the inhibition of TF function by most of the monoclonal antibodies assessed was dependent upon the dissociation of the TF/VIIa complex that is rapidly formed when TF contacts plasma. One monoclonal antibody, TF8-5G9, was capable of inhibiting the TF/VIIa complex without dissociation of the complex, thus providing an immediate anticoagulant effect in plasma, as disclosed in WO 96/40,921.
Targeted clotting factors exhibit both a medium molecular weight range (about 45,000 to 160,000) and a relatively high normal plasma concentration (at least 0.01 micromol/L).
One persistent concern with all available anti-thrombotic agents is the risk of overdose and therefore of excessive and life-threatening bleeding. Most current antithrombotic agents therefore warrant close monitoring of the patient.
Thus, there is a need for efficient compounds for the treatment of coagulation disorders, which cannot be overdosed, require no monitoring and are free from bleeding problems. For a therapeutic agent based on antibodies, the ideal compound would be a human antibody with full anticoagulant efficacy that does not induce immunogenicity.
Factor VIII is a protein providing important_coagulant cofactor activity and is one of human clotting factors with a rather high molecular weight (265,000) and a very low normal plasma concentration (0.0007 micromol./liter). With its 2,332 amino-acid residues, Factor VIII is one of the longest known polypeptide chains and is synthesized in the liver, the spleen and the placenta. Its gene has been shown to include 186,000 nucleotides.
Factor VIII circulates as inactive plasma protein. Factors V and VIII are homologous proteins sharing a common structural configuration of triplicated A domains and duplicated C domains with structurally divergent B domains connecting the A2 and A3 domains. Factor VIII circulates in a multiplicity of fragmented species in a tightly associated complex with von Willebrand factor at a concentration of 1 nmol/L. Factor VIII activation occurs by a cleavage between the A1 and A2 domains, resulting in the unstable heterotrimeric Factor VIIIa molecule. Factor VIIIa binds tightly to membranes that contain acidic phospholipids. Factor VIII contains a phospholipid binding site in the C2 domain, between amino-acids 2302 and 2332, according to Arai et al. in J. Clin. Invest. (1989) 83:1978. Within the same Factor VIII region, there is also a von Willebrand factor binding site acting in conjunction with amino-acid residues 1645-1689 in the A3 domain according to Shima et al. in Throm. Haemost. (1993) 69:240 and J. Biol. Chem. (1994) 269:11601.2 
Polyclonal antibodies inhibiting the co-factor activity of Factor VIII have been classified as type I or type II inhibitors according to their capacity to inhibit Factor VIII either completely (type I) or only partially (type II). According to Gawryl et al., Blood (1982) 60:1103-9, the reduced inactivation of Factor VIII by human type II autoantibodies is believed to be due to a steric effect of von Willebrand factor. Monoclonal antibodies_are not mentioned and, to date, no therapeutic use was made of such type II inhibitors. Biggs et al., Br. J. Haematol. (1972) 23:137 previously provided an interpretation derived from data obtained by using human polyclonal antibodies, that a type II inhibitory pattern could be related to low affinity. B. Ly et al., Scandinavian Journal of Haematology (1982), 28:132-140 discloses polyclonal antibodies to Factor VIII which most often belong to the IgG class both in hemophiliacs developing alloantibodies and in the more rare patients having autoantibodies against their own Factor VIII. These polyclonal antibodies partially inactivate Factor VIII activity like the antibodies described in Biggs et al. (1972) and Hoyer et al. (1982), This document again fails to mention whether monoclonal antibodies can reproduce the pattern of Factor VIII inactivation shown by patient's polyclonal antibodies. Again, no monoclonal antibodies are mentioned.
European patent applications EP-A-123,945, EP-A-152,746 and EP-A-432,134 all disclose monoclonal antibodies produced by hybridoma cell lines and having a specific reactivity pattern with Factor VIIIc polypeptide fragments. These monoclonal antibodies are said to be useful for detecting the presence of Factor VIIIc and related polypeptides in plasma by immunoassay techniques, but a therapeutic potential use is not suggested in these documents.
J. Battle et al., Annals of Hematology (1997) 75:111-115, discloses a polyclonal alloantibody from a patient with severe von Willebrand disease showing, alike a rabbit polyclonal antibody against von Willebrand factor, a partial inhibitory activity to plasma Factor VIII. These polyclonal anti-Factor VIII antibodies therefore inactivate Factor VIII following a pattern similar to anti-Factor VIII type II antibodies found in patients with hemophilia A (Gawryl et al., Blood (1982) 60:1103-9). However, Factor VIII antibodies were not detected in the said human alloantibody, thus suggesting that it was a non-specific inhibition.
J. Ingerslev et al., Clinica Chimica Acta (1988) 174:65-82 discloses a series of murine monoclonal antibodies against human von Willebrand factor: two of them, belonging to the immunoglobulin isotype IgG1, exhibit an extremely low (1.3 BU/mg immunoglobulin) inhibition of Factor VIII as shown in table I of said document. By comparison, human monoclonal antibody BO2C11, derived from a hemophilia A patient with inhibitor, has a specific activity of 7,000 BU/mg protein (Jacquemin et al. Blood, (1998) 92:496-506). This indicates that administration of antibodies as described by Ingerslev to an animal or a human being would not affect Factor VIII activity, unless an extremely high amount of antibody (hundreds of mg/ml) was present in plasma. The authors do not disclose whether when used in large excess these antibodies exhibit inhibitory activity like type I or type II (i.e. partial inactivation) polyclonal human Factor VIII inhibitor, such as described in Gawryl et al., Blood (1982) 60:1103-9.
Maraganore et al., Circulation (1992) 86:413, showed that a synthetic 12-aminoacid peptide corresponding to residues 1675-1686 of Factor VIII inhibits cleavage by thrombin of the heavy chain required for the activation of the procoagulant activity of Factor VIII and also of the light chain required to dissociate Factor VIII from von Willebrand factor and that tyrosine sulfation of said peptide potentiates its recognition by Factor VIII.
O'Brien et al., J. Clin. Invest. (1988) 82:206-211 describes obtaining an animal model for hemophilia A by infusion of human anti-Factor VIII antibody in rabbits. According to WO 95/01570, antibodies against the light chain of human or porcine Factor VIIIc were produced in a first animal and subsequently a temporary hemophilia was induced in a second animal by means of the purified monospecific antibody obtained. U.S. Pat. No. 5,804,159 also discloses inducing a temporary clotting disorder in a mammal by means of an anti-plasma antibody preparation acting on several blood coagulation factors, e.g. a preparation comprising antibodies against human von Willebrand factor and Factor VIII, or against Factor VIII/von Willebrand factor-complex, or against procoagulants, anticoagulants, clot structure factors, fibrinolysis factors and phospholipids.
However, none of the above-mentioned antibodies compounds involving Factor VIII have been described for therapeutic purposes. In fact there is a prejudice among those skilled in the art against investigating anti-Factor VIII antibodies for anti-thrombotic therapy because it is assumed that, a deficiency in Factor VIII being the cause of hemophilia A, such antibodies would induce a bleeding state.
WO97/26010 discloses monoclonal antibodies having self-limiting neutralizing activity against a coagulation factor which are useful in pharmaceutical compositions for thrombotic disorders._Self-limiting neutralizing activity in this document is defined as the activity of an antibody that binds to a human coagulation factor and inhibits thrombosis in a manner such that limited modulation of coagulation is produced. Limited modulation of coagulation in turn is defined as an increase in clotting time as measured by prolongation of the activated partial thromboplastin time (aPTT) where plasma remains clottable with aPTT reaching a maximal value, preferably 35 to 100 seconds, despite increasing concentrations of the monoclonal antibody. APTT is thus used as the primary criterion for the evaluation of efficacy versus bleeding liability of antithrombotic agents.
More particularly, this document demonstrates that a sheep polyclonal to Factor VIII (SAF8C-IG, purchased from Affinity Biologicals) induces a self-limiting prolongation of aPTT (the aPTT increased to a maximum of about 65 seconds). We have demonstrated, however, that SAF8C-IG totally inhibits the activity of human Factor VIII (see FIG. 10), i.e. is a type I inhibitor in the classification of Gawryl et al., Blood (1982) 60: 1103-9. This demonstrates that a limited increase in clotting time up to a certain maximum value is not necessarily correlated with partial inactivation of a clotting factor, and far less to a decrease in the risk of bleeding. For instance, it is well known that patients with a complete deficit of coagulation factors have a limited prolongation of aPTT, usually in the area of 60 to 100 seconds, but are nevertheless exposed to a dramatic risk of bleeding (Hathaway et al. Am J Clin Pathol (1979) 71: 22-25, and Hoffmann et al. Thromb Haemostas (1978) 39: 640-645).
Conversely, it is well known that a prolonged APTT does not provide a valid parameter of the reduction of thrombosis risk. Notably, deficiency in Factor XII, another coagulation factor of the intrinsic coagulation pathway results in APTT prolonged up to 6-fold (Hathaway et al. Am J Clin Pathol (1979) 71: 22-25, and; Hoffmann et al. Thromb Haemostas (1978) 39: 640-645). However, a significant number of patients with this deficiency have experienced myocardial infarction or thromboembolism, demonstrating the lack of protection from thrombotic disease in patient deficient in Factor XII, despite important prolongation of the APTT (McPherson R A Am J Clin Pathol (1977) 68: 420, and; Glueck H I et al. Ann Intern Med (1966) 64:390).
Jacquemin et al. in Blood (1998) 92:496-506 refers to a Factor VIII-specific human IgG4 monoclonal antibody (BO2C11) produced by a cell line derived from the memory B-cell repertoire of a hemophilia A patient with inhibitors. BO2C11 is said to recognize the C2 domain of Factor VIII and to inhibit its binding to both von
Willebrand factor and phospholipids. It is said to completely inhibit the procoagulant activity of native and activated Factor VIII with a specific activity of 7,000 Bethesda units/mg. The present inventors have further shown that BO2C11, while totally inhibiting the activity of human Factor VIII, provides a prolongation of about 110 seconds in clotting time as measured by aPTT, which again demonstrates that an increase in clotting time up to a certain maximum value is not necessarily correlated to partial inactivation of a coagulation factor. Such a reduction of Factor VIII levels would expose the patient to severe risks of bleeding, like in patients with severe hemophilia A (Levine P H Ann NY Acad Sci (1975) 240:201; Gilbert M S Mount Sinai J Med (1977) 44: 339).